Date of Award

2013

Document Type

Dissertation

Degree Name

Doctor of Philosophy (PhD)

Department

Educational Leadership

First Advisor

Whitaker, Todd

Abstract

This study used existing third-grade ISTEP+ data from a cohort of Evansville Vanderburgh School Corporation students to evaluate various non-experimental methods for estimating teacher effects on student test scores. The study considered the impact that first-, second-, and third-grade teachers had on student performance as measured by the Indiana standardized test in third grade by looking for recognizable patterns of success based on teacher assignment in a cohort of 350 students. By making the assumption of randomness in assigning students to teachers and controlling for student transience, demographics, and teacher movement, the variances of mean ISTEP+ scores were examined to determine and quantify differences based on teacher links. Descriptive statistics summarized possible patterns of success based on teacher links for the cohort as a whole and each school individually by grade level. Differences among teachers by grade level were examined by using an ANOVA model. Regression analysis was used to probe patterns of achievement based on teacher combinations as well as the predictability of ISTEP+ scores based on first-, second-, or third-grade teachers.

Share

COinS